On Friday February 11th 2011, the Obama administration released its plan for reforming the US housing market. The bulk of the plan deals with the future of the Government-Sponsored Enterprises (GSEs) – FNMA, FHMLC and GNMA. As part of the release, the administration reiterated their commitment to winding down Fannie Mae and Freddie Mac and proposed several potential future models for housing finance in the US. All three models are dependent on the return of private capital to the housing market to replace the GSEs, which currently insure or guarantee more than nine out of ten loans made in the wake of the credit crisis.
As we have expressed in previous postings (most notably “Can You Hear Us Now, Mr. Geithner?” ), data transparency in the mortgage markets is of utmost importance, and the reluctance of the GSEs to provide such data has been a thorn in the industry’s side for quite some time. A market shift from public to private financing would have many benefits, not least of which would be the reduction of federal exposure to the ups and downs of the mortgage market. But of equal importance would be the increased availability of loan-level data for mortgage-backed securities, which would allow investors to truly understand the risk composition of their investments. 1010data applauds the intent of the administration’s plan for reform, and looks forward to seeing the results of its implementation.
An article in the Star Tribune, Cloud of Secrecy Surrounds Credit Scores, revealed new credit scoring methods beyond FICO, the standard that many people are familiar with. It strikes a negative and alarming tone, which is not surprising given the title, and says that consumers should be concerned about the use of their data and lack of transparency into the new methods. Here are a few words from it:
Banks and credit card issuers are digging deeper into people’s lives, using powerful new tools. Beyond the credit score, they are looking at where you live, how often you switch jobs or telephone lines, whether you get paid with direct deposit and other factors to decide whether you deserve a loan. And unlike the well-known FICO score, which can be easily monitored online, most of the new scoring models have been developed solely for the use of lenders. As a result, would-be borrowers may be denied credit based on factors they never knew were relevant.
Looking beyond the main point, which presents a valid concern, the article casts the industry – more specifically lenders, data providers and credit reporting bureaus – as villains and consumers as hapless victims.
We would never argue that consumer privacy is not important – and consumers should of course understand how data about them is accessed and used, and how this can affect lending decisions. At the same time, it is important to point out that transparency can work in many ways – increased industry access to data can often help consumers – and not just those who with the pristine credit records.
E.g. as we said in our article Can You Hear us Now Mr. Geithner, it is important for those in the mortgage-backed security business to have access to certain types of data. This helps them in their work, but it also can help consumers and directly affect things like availability of capital for lending and mortgage interest rates (and be done in a way that does not jeopardize privacy concerns). We wrote in that post about how lack of access to consumer loan level data from Fannie Mae and Freddie Mac can have an adverse impact on the price and availability of mortgage credit for consumers. I encourage you to visit the link and read this post, which explains in detail why this is true.
In summary, it is simplistic to paint the industry as villains and consumers as hapless victims when access to consumer data is the topic; transparency can work many ways, and industry access to data can often help consumers.
1010data is at its heart a technology company. However our aim is to not just provide more and better “bells and whistles.” Our hope is that our service can be used to improve the ways our customers make decisions and adapt to marketplace dynamics.
Looking beyond business decision making and profits, there can be a higher principal at play. We are a strong believer in the ability of technology to provide access to information that can serve the public good. This is particularly relevant and true as it relates to our work in the asset backed securities arena. E.g. we have blogged about the benefits of achieving greater transparency in ABS markets (see European Non-vacation: Transparency and the State of the Union and Can You Hear Us Now, Tim Geithner?).
For all of the above reasons, we were pleased to announce our work with the American Securitization Forum in issuing a series of free reports that provide data from some of the industry’s leading vendors. As the press release says, the reports make available timely summary information on key segments of the mortgage-backed securities (MBS) market.
This truly is a first for the industry. Put simply, the reports make more data available to more people. We understand that there can be such a thing as “too much information,” and the goal is not to provide data by the pound just because we can. Rather, through the data, a wider range of people can get key insights into the state of the non-agency (i.e. non- Fannie Mae or Freddie Mac) MBS market, including segments such as subprime, Alt-A, and Prime MBS. The report strives to provide the market with a small slice of the kind of detailed insight our customers enjoy.
Already, since we made the announcement last week and issued the first report, people have written articles that tap into the data. Housingwire came out with this article, which includes this excerpt.
Data released Tuesday by 1010data, in conjunction with the American Securitization Forum, CoreLogic and Equifax, confirmed what CoreLogic economist Mark Fleming mentioned during an interview with Housingwire one month ago — Alt-A loans are the new subprime.
The information was picked up by others, for example see Irvine Housing Blog.
The first report is on our website, click here to view it.
CNBC’s Steve Liesman reports that big changes may be in store for the banking system in his story, Who Would Finance Mortgages If Fannie, Freddie Disbanded?
We all understand that, while banks write mortgages, it’s the mortgage-backed securities (MBS) market that helps make the current system work. If you take away the guarantee that comes with an agency mortgage (i.e. Fannie Mae or Freddie Mac) the system can fall apart. And, as Steve notes in his article:
…without the guarantees, experts say, there would be no securitization, no capital from the rest of the world for long-term fixed rate mortgages and banks would have to hold on to them.
Setting aside the issue of desperately needed structural change at the GSE’s, we have to agree with the statement and these experts.
Of course, without securitization, consumer interest rates would revert to where they otherwise most likely would be… several points higher. What this would mean is lower home values at less affordable interest rates.
I think part of the issue is that the government has taken a paternalistic stance towards home ownership and firmly believes that, as a society, home ownership brings greater social stability. But if you look at Europe, where there is no government-sponsored entity to promote housing affordability, home ownership is much lower. They have a society of renters.
The question here is really whether a nation of renters is worse than continuing to allow the government to subsidize home ownership for all with our tax dollars (and absorb the subsequent massive losses).
What do you think? Should we end government guarantees?
As I mentioned in my last post, 1010data exhibited at the Global ABS 2010 conference in London earlier this month. The show covered a wide range of topics related to securitization. We had many productive conversations, and gained a better understanding of issues that are important for the European market.
As readers of this blog know, one of the hot buttons for 1010data is transparency, or rather the lack of it. You may recall the recent post about our letter-writing campaign (Can You Hear us Now, Tim Geithner?), which aimed to drive improved transparency into the mortgage pools of the major agencies here in the U.S. One thing we learned at Global ABS that made us feel at least a little better is that, as opaque as it can be here in the U.S., there is even less transparency in Europe.
There are probably many good reasons for this. There are no publicly-backed lending institutions comparable to Freddie Mae and Fannie Mac for the European Union – or even in individual E.U. countries. There is less homogeneity across Europe; the laws that govern banking can be more complex and vary from country to country. Rules about privacy of data can vary as well, making it harder to make information that can lead to improved transparency more widely available.
The upshot of all of this is that investors in mortgage-backed securities really have nowhere to turn when it comes to having the data and analytical tools that can help them better understand the risk inherent in mortgage pools in Europe. We know there is a desire for access to this type of data, as we have spoken directly with European investors about this.
There is a ray of hope on the horizon, and some in Europe are pushing for improved transparency. E.g., the Bank of England is making waves about the release of data – see this article which ran earlier this week in the Daily Mail, a U.K. publication. It says:
Between now and the end of 2012, British banks need to refinance-750 billion of loans. The Bank wants as much of that as possible to come from domestic investment institutions with low borrowings, rather than from highly-geared, foreign investors making short-term investments.
But the Bank fears this will not happen if domestic investors lack confidence in the reporting of financial institutions.
A spokesman for the British Bankers’ Association said: ‘ British banks fully understand the need for transparency and are committed to the highest possible accounting standards.’
Please pardon the interruption from our regularly scheduled topics on BI and analytics and allow us to vent a bit.
As those who read this blog probably know, 1010data, the company behind The Future of Data, does lots of work in the mortgage industry. We provide mortgage and asset-backed securities participants with data, analytics and ad hoc query tools to recognize and respond to changing market dynamics. As such, we think that our unique perch gives us an excellent perspective on the mortgage business – an area that has undergone much scrutiny in light of the devastation caused by the mortgage market meltdown over the past couple of years.
Despite all the hard work of the Obama administration, the Bush administration before it, and the banking industry to fix the mess, and the progress achieved so far, there is one area that has been ignored. It may a be a bit harder to understand, and doesn’t come with the broad strokes or headline-grabbing potential of a bailout. Yet there exists a relatively simple way to drive specific and concrete improvements in the mortgage business and housing market – and this fix has been ignored by Treasury Secretary Tim Geithner, as well as members of the House Financial Oversight Committee, despite our pleas.
Let us explain. To do their jobs well, traders of mortgage-backed securities need to understand the mortgages bundled into these securities. Mortgage securities can have vastly different risk profiles despite looking very similar when data about each of the loans in the pool are available only at an aggregate level.
Today, widely available technology lets you get down to the lowest level of detail for non-Agency mortgages – i.e. to explore individual loan and borrower data (without revealing identifying information). This sheds light on the composition and risk profile, of the mortgage pool at hand and helps traders and risk managers adjust strategies based on the likelihood of loan defaults and prepayments, and helps loan originators like banks track the effects on loan performance of changes in underwriting standards as they seek to improve their practices.
Unfortunately, for no good reason that we can see, this type of detailed loan-level data continues to be withheld from the investment community, and from the American public at large (whose tax dollars, in fact, pay for the Agencies). The Agencies provide liquidity for three quarters of the residential mortgages in America, yet they provide data only at an aggregated, pool level – frustrating efforts of the mortgage securities industry to better understand the health of these pools. Improved transparency via loan-level data is critical for mortgage bond investors, and has a direct impact on the price and availability of mortgage credit for consumers.
Months ago we wrote letters to Tim Geithner and members of the aforementioned House committee. We even offered our services and technology, at no cost to the Agencies, to make the data available. We have not received a single response. We are hoping that if we (and you) call, Mr. Geithner would pick up the phone and hear us now.
Please, take a moment to contact the Department of The Treasury to share your thoughts.
My readers know about my frequent call for increased transparency in the mortgage securities industry and I’m pleased to say that today we announced a partnership that works to move that goal ever closer.
S&P Valuation and Risk Strategies, an independent and analytically separate business unit within Standard & Poor’s, announced an agreement to offer new investor analytics for U.S. residential mortgage-backed securities (RMBS) via 1010data. Together we bring loan-level performance data on subprime, alternative-A, prime jumbo and additional collateral type loans to our joint customers. S&P’s Jonathan Reeve said that working with 1010data, makes it easier for them to deliverer a comprehensive risk management solutions, allowing investors to quickly dig down into the foundational components of their portfolios in a way that was never before possible.
I was glad to see John Hintze’s article, Goldman CDO: The Loan Data Was There for All to See in Securities Industry News. It was extremely refreshing to see a journalist make the point that many in the industry have been waiting to see published since the news broke about the Abacus deal.
According to the article:
…detailed and updated data about these assets-mortgage loans populating 90 mortgage-backed securities that made up the $2 billion CDO called ABACUS 2007-ACI-was there for all to analyze…
The fact that the data was out there for investors to analyze is extremely important. It means buyers of the Goldman bonds had enough information to determine that the deal could suffer from extreme losses. Some of the controversy around this CDO revolves around the parties involved in selecting the bonds in the deal. Leaving this particular issue aside, the mere fact that the data was there for all investors to analyze is a huge point. If the investors had been doing their jobs properly, they could have looked right at the collateral and determined the quality of the bonds themselves.
There was also a related story in the Wall Street Journal yesterday: Toxic CDOs Beset FDIC as Banks Fail. The article reports:
The Federal Deposit Insurance Corp., and by extension the U.S. taxpayer, owns more than 250 collateralized debt obligations that were purchased by small institutions that later failed. Although the bonds have a book value of more than $400 million, they are a headache for the agency as it grapples with the toxic assets flowing from many banks around the country…The FDIC‘s focus on CDOs comes at a time when the financial instruments are being scrutinized by regulators and prosecutors. Several Wall Street firms, including Goldman Sachs Group Inc. and Morgan Stanley, have attracted particular attention in recent weeks for what they told investors about the nature of the CDOs when the initially sold them.
Here too, the data is there for all to see.
One way the current SEC proposal attempts to bring transparency to the ABS/MBS market is through the release of loan level data. Conceptually this is a move in the right direction, and I don’t think anyone is going to argue against the need for transparency. However, there is one BIG issue. Footnote 230 on page 110 lays out where the proposal falls short: “Fannie Mae and Freddie Mac are government sponsored enterprises (GSE’s) that purchase mortgage loans and issue or guarantee mortgage-backed securities (MBS). MBS issued or guaranteed by these GSEs have been and continue to be exempt from registration under the Securities Act and reporting under the Securities Exchange Act. As a result, only non-GSE ABS, or so called “private label” ABS, will be required to comply with the new rules. For more information regarding the GSEs, see Task Force on Mortgage-Backed Securities Disclosure, “Staff Report: Enhancing Disclosure in the Mortgage-Backed Securities Markets” (Jan. 2003) available on our Web site at http://www.sec.gov/news/studies/mortgagebacked.htm.”
Additionally, the release of standardized loan level data for loans in non-agency MBS is a nice to have, not a need to have. I say this because this data already exists in the market and is available from multiple providers. It is NOT available for loans in Agency MBS. After everything we’ve been through, how can the government continue to allow the GSE’s to shield the loans in Agency MBS from the market? As a taxpayer, I find that frustrating.